Language…
24 users online:  Alex, anonimzwx, DanMario24YT, Dennsen86, derv82, Domokun007, drkrdnk, GRIMMKIN, h.carrell, iRhyiku, Joaokk19990, Maniek, Maw, Metal-Yoshi94, Nayfal,  NopeContest, Red2010, Serena, ShadowMistressYuko, Silver_Revolver, sinseiga, steelsburg, Tulip Time Scholarship Games, yno14jax - Guests: 291 - Bots: 414
Users: 64,795 (2,374 active)
Latest user: mathew

Posts by Tamaki

Tamaki's Profile → Posts

(restricted)
File Name: Final Fantasy 4 - Dancing Doll Calcobrena
Submitted: by Knecht Ruprecht🎅
Authors: oL7G5poF4c
Insert Size: 0x0220 bytes
Type: Song
Sample Usage: None
Source: Port
Duration:
Featured: No
Description: A Japanese port of Final Fantasy 4 - Dancing Doll Calcobrena
As our submission guidelines state:

Originally posted by Music Submission Guidelines
If you are not the author of the submission, please provide proof in your submission that you have permission to upload it, or in the description field of your submission. Otherwise, have it readily available if a moderator inquires.
Uploading another person's work without their consent is expressly forbidden.


You didn't provide any proof that you were authorized to submit this on behalf of the author, so I'm removing this. Please don't do it again, okay?
I'm... speechless. Thank you very much friends! I wasn't expecting so many people voting for me, so I truly appreciate it. #smrpg{<3}
And congratulations to the other winners too! I see many well deserved trophies in there. #tb{:p}
(restricted)









There was no change in the note data, only in the submissions' information. This was already discussed with the author through PMs, but for the general public: if you want to update tags, description or any type of information of your already accepted submissions, feel free to message a music moderator! We'll gladly edit it so that you don't lose the comments and ratings. Plus, it's faster this way for both parties. #tb{:p}
(restricted)
Can you also link your samples and an spc file? Clicking has multiple causes, though the most common one is a badly-looped brr file. The only way to fix it is to convert the file to .wav, then loop it again manually so it repeats seamlessly.
Okay, I think I see the issue now. It seems like the song isn't looping properly because some channels are slightly shorter than the others. To fix this, go to the end of channels #0, #1, #3 and #6 and change the r16 to r8^32.

Also, this might not be the case for your port, but it's a good practice to set the GAIN byte (the middle one) in your custom instruments to values $80 or higher, as lower values will cause clicks if you use remote commands with event 3. So, instead of:

Code
"Rock Organ.brr" 		$ff $e0 $00 $02 $1f


You use (for every instrument):

Code
"Rock Organ.brr" 		$ff $e0 $80 $02 $1f
Really? That's odd. I tried it and it sounds fine for me. Here's the .txt, in case you need it.
(restricted)
File Name: Newer Super Mario Bros. Wii - Pumpkin
Added:
Authors: Sayuri
Insert Size: 0x52E bytes
Type: Song
Sample Usage: None
Source: Port
Duration: 1:17
Featured: No
Description: The haunted-sounding song from the popular ROM hack of New Super Mario Bros. Wii, Newer Super Mario Bros. Wii, that plays in pumpkin-themed levels. Good for haunted levels.

Obsoleted by bebn legg's version, as it's more accurate to the source material and it's also lighter in insert size.

Archived here.
File Name: Mega Man ZX - Pallida Mors
Submitted: by Ultima
Authors: Ultima
Insert Size: 0x1F20 bytes
Type: Song
Sample Usage: Many
Source: Remix
Duration: 3:30
Featured: No
Description: And here's something I don't think anyone here would expect: a remix (no, it's not an arrangement this time, I mean an actual remix) of the final boss theme from Megaman ZX, Pallida Mors. This one was made for a certain competition for the Chiptune Cafe Discord server, and since it's been a while that the competition ended, I think it's a good enough opportunity to release this now haha

Anyway, this one is very different to the original in many different ways, but the biggest change specifically is the extra section I've added at 2:47. Since this was made mainly thinking about the competition itself and not specifically for SMW, it doesn't loop like the original does. Despite that tho, I have made a looped version for it, and honestly, the song is long enough by it's own that it fits it's purpose well enough to work even without the loop.

There's also optimized versions for each version as well; these don't have the echo and removes a few SFX samples, like Yoshi bongos and the thunder SFX, but I feel like it could be done better than this, I just couldn't be bothered really lmao

But anyway... this one I feel like it could only really work in a final boss setting, but the looped version is there if you want to use it for a level pretty much. Not much else to say here, just have fun :P

SPC

I really like this port, and it sounds like a ton of effort was put into it! I've got no complaints on the way it sounds, but I did find many big optimization problems in the MML file that could reduce the insert size significantly if you fix them. I'm going to mention the errors that stick out the most, but keep in mind that there are more repeating patterns in the file.



- In channel #1 you have a series of label loops that you repeat in the same order right away (Image). This can be replaced with a super loop for that big section. Additionally, doing this will allow you to loop the repeating patterns inside the old label loops, which you can find very easily with the Smart Highlighting feature in Notepad++ that Maxodex mentioned here. Channel #0 also has this issue, so make sure to take a look at it as well.

- Lines 227 and 247 are identical, so you can use a label loop there. Image.

- In channel #0, for label loop number 11 and 12, you have the same three notes repeated four times inside. Image.

- For channel #2 there are two identical lines that can be label looped (Image). A few lines below that is another case of repeating notes inside a loop (Image). It's worth mentioning that this pattern is repeated many times in that channel, so you can save more with label loops.

- These lines in channel #5 can be condensed in a super loop. Image.

As I said before, these are the mistakes that stick out the most, but they're certainly not the only ones. I encourage you to take a closer look at the MML to find more to make this port smaller in insert size!



If you have any questions, feel free to DM me on Discord! I'm available most of the time. #tb{:p}
(restricted)
File Name: SaGa Frontier - Battle 1
Submitted: by Pat
Authors: Pat
Insert Size: 0x75F bytes
Type: Song
Sample Usage: Many
Source: Port
Duration: 1:12
Featured: No
Description: Battle theme from SaGa Frontier that has an antagonistic vibe to it. Fits in tense levels, like castles.

Thanks to Samantha for help with samples and looping!

The MIDI I used came from here: https://fantasyanime.com/saga/sagamusic.htm

SPC

This is a good attempt! However, I have to reject it due to the following issues:


1. Most instruments are out of tune. Some are slightly flat (lower pitch) and others are slightly sharp (higher pitch). To not make this very long, I'm going to list the samples and their correct tuning values.

- Slap Bass: $05 $00.

- Guitar: $02 $FF, or $06 $00 if you lower the octave (I recommend the latter).

- Strings: $01 $09.

- Lead: $06 $00.

- Kick: $0A $1D.

- Snare: $0A $9F.

- Closed Hi-Hat: $03 $BF.

The Rock Organ sample is tuned correctly.

Also, I tuned the samples to match the notes and octaves you've assigned for them, so there should be no need to modify anything else.

2. The attack value of the Guitar and the Lead isn't the best, and they drastically change the feeling of the song. In this case, all the samples should have an attack of F if you're striving for accuracy.

3. You're missing three percussion samples: the Open Hi-Hat, the Crash Cymbal and the Tom. Because you don't have a Tom, your Snare sample is playing notes it's not supposed to play and it doesn't sound good. The missing Open Hi-Hat and Crash Cymbal samples make the percussion and the port as a whole sound empty.

4. Touching on the last point, you're also missing some important harmony channels. I know you can't cram every single channel from the original into the SNES, but the reason this is a problem is because sometimes you have up to three empty channels that could be used to add more body to the port. In addition, by adding the Open Hi-Hat sample, you can free the Noise channel entirely, which gives you room to include the second Strings channel in the intro and the second Rock Organ channel after it.

5. When using Legato ($F4 $01), you have to make sure to disable it inside the last note that you want it to affect. The reason for this is that, sometimes, when you turn it off outside the last note, a click will occur when the first note without Legato plays. It's not a big issue, but it's a good practice to disable it properly.

For example, in the first channel you have $F4$01 b8. > c+8. $F4$01 d8, which would look like this if you were to fix the placement of Legato: $F4$01 b8. > c+8 $F4$01 ^16 d8. If the original note doesn't have a tie or a dot, you can cut it in half to achieve the same result: c16 $F4 $01 ^16 (the original was a c8).


To close this, I'd like to touch a little on the optimization. It's pretty decent overall! However, I do want to mention a few things. First off, although it's true that super loops use less bytes than regular loops, it's much more efficient to use label loops if the pattern of notes repeat afterwards. Channel #1, #4 and #6 have plenty of examples of this. Moreover, you can have super loops inside regular and label loops to save more space. While this port won't benefit much from this, it's still good to keep it in mind. Take this part in #1 for example; you can use a label loop to encompass the notes highlighted in red and then call them below. Lastly, you missed a couple of long rests in #5 where you can use (500).


I hope you can fix the issues I described here because your port has a lot potential! In case you have any questions, feel free to contact me on Discord! #tb{:j}
I'd love to add comments, but time isn't on my side right now, unfortunately. #ab{>_<}

1. Alpha
2. oboro
3. gourmetrace
4. remix_9
5. Operation 2
6. Route 113
7. studiopolis1
8. Dive
9. Splash Hill Zone 1
10. cosmicangel
11. kyushu
12. PokeFlute v2
13. mm2gb_title
14. test29
15. Pyramid
16. 72ho6_staircase_whip



Originally posted by xfix
studiopolis1 (disappointed there is no voice acting, but I mean, it's SNES, and I think the track is still alright without it, section at 0:24 seems somehow off to me)

The rules of the contest didn't allow custom samples. I'm sure the porter could have added them if they were an option. #tb{:p}
File Name: Super Mario Bros 3 - Ocean Land
Submitted: by EDIT3333
Authors: EDIT3333
Insert Size: 0x0366 bytes
Type: Song
Sample Usage: Many
Source: Remix
Duration: 0:12
Featured: No
Description: I changed the pitch of the melody and added another channel

(Refer to the previous post for the SPC, as it's identical.)

From what I can tell, the submission is exactly the same as last time, with the exception of having the "../" part removed from one of the samples. Nothing bebn mentioned was addressed; for this reason, I'll have to reject the port.

Please don't submit this again unless you've fixed the two major issues detailed in the first post, okay?
File Name: Playstation 2 Startup
Submitted: by Samantha
Authors: Samantha
Insert Size: 0x0040 bytes
Type: Song
Sample Usage: Many
Source: Port
Duration: 0:12
Featured: No
Description: The iconic Playstation 2 startup sound

SPC

Since this is only a note with a really long sample, I have to be stricter when deciding whether or not it's acceptable in the music section. With that said, this needs work on the tuning of the sample, as it's very low-pitched compared to the original. If you transpose the note by 5 semitones up, you might notice that it's closer to the original, but it's not quite correct, so you'll need to toy with the tuning a little more.

This also brings another issue: the audio file was downsampled so much that it's full of artifacts when the pitch is raised to sound closer to the source. While this isn't exactly a removal reason, you might want to consider toning down the downsampling a bit if you remake the sample so you get a clearer sound.
File Name: Gradius III - Departure for Space
Submitted: by Infinity
Authors: Infinity
Insert Size: 0x0896 bytes
Type: Song
Sample Usage: None
Source: Port
Duration: 0:30
Featured: No
Description: Song that Departure for Space from Gradius III. Good a for space and athletic levels.

SPC

This port has several issues, which unfortunately prompt me to reject it.

To start off, your tempo is t91; an extremely high value that has the side effect of slowing down the music engine, not only when the music is playing, but also when the "Time is running out!" sound effect plays. This in turn causes certain sound effects such as the Spring Board to lag, too. Please avoid using values over t80 to avoid this, as our engine isn't very well optimized to handle anything higher. If you want, you can halve your tempo and then follow the instructions from here to also halve the notes.

Second, your port is missing quite a few channels from the original song. Examples of this include the chords at the beginning (0:01 to 0:09) and its pseudo echo channel, as well as the other two choir channels for the section that follows it. While you do dedicate two channels to the intro, they're playing the sames notes with different instruments, while in the original they're the complete opposite: different notes and the same instrument.

Third, channel #3 goes slightly out of sync as the song progresses, which is most likely caused by the seemingly random r192s found throughout the channel. It's also worth mentioning that all the fast notes are missing, so it sounds very simple and empty.

Fourth, your mixing could be a lot better. For instance, the most prominent channel in the original is the bass, which is very quiet in your port because of all the other channels overpowering it. In general, it is not a very good idea to give everything the same volume values (v220) because the samples vary on their loudness, so you have to adjust accordingly.

And last but not least, the optimization could be improved. You can use [[ ]] in conjunction with label loops to better optimize #7, as the following lines repeat quite often (Image). Also, in #3 you have many repeating chunks of notes (Image). And speaking of that channel, if you had consistent rests, you'd be able to optimize it much further, especially at the beginning.

All in all, this port isn't all that bad, but it certainly needs more work before I'll be able to accept it in the music section. Since fixing all of this might be more work than it's worth, I suggest you to find a better MIDI of the song and start over, because I can tell at least half of these issues aren't directly your fault.
File Name: Taps
Submitted: by MegaNerd14
Authors: MegaNerd14
Insert Size: 0x00A6 bytes
Type: Song
Sample Usage: Many
Source: Port
Duration: 0:52
Featured: No
Description: *salutes*

SPC

The main problem I have with this one is how grating to the ears the Trumpets are. Even though the real song doesn't have vibrato, this version would benefit from it because the loop of the sample is very small and sounds dull as-is.

Aside from that, the second channel constantly amplifies the first one because of the long notes, which makes the first problem even more prominent. A way to fix this would be to slightly change the pitch of the second channel with the $EE command; $EE $0A would be my suggestion, but you can change the value until you find a sweet spot for you.

Another suggestion I wanted to make is to use a slower Attack for your ADSR to mimic the real trumpet a bit better. You could even have varying Attack speeds for the shorter notes (faster Attack) and the long ones (slower Attack) to spice things up a bit.

Lastly, as a minor note because there are many versions of Taps, your port is three semitones lower than the version I know, and the descending notes starting at 0:32 have incorrect lengths, but I'm wiling to let these two points slide if you can specify the version you're basing your port on.
(restricted)